Table Of ContentOSSE Preprint No. 6
The Oriented Scintillation
Spectrometer Experiment
Instrument Description
To appear in Astrophysical Journal Supplements
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it
does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
1993 2. REPORT TYPE 00-00-1993 to 00-00-1993
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
The Oriented Scintillation Spectrometer Experiment Instrument
5b. GRANT NUMBER
Description
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
Naval Research Laboratory,E.O. Hulburt Center for Space REPORT NUMBER
Research,4555 Overlook Avenue, SW,Washington,DC,20375
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE 67
unclassified unclassified unclassified
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
The Oriented Scintillation Spectrometer Experiment
Instrument Description
W. N. Johnson, R. L. Kinzer, J. D. Kurfess, M. S. Strickman
E. O. Hulburt Center for Space Research, Naval Research Lab, Washington DC 20375
W. R. Purcell, D. A. Grabelsky, M. P. Ulmer
Northwestern University, Evanston, IL
D. A. Hillis
Ball Aerospace Systems Group, Boulder, CO
G. V. Jung and R. A. Cameron
Universities Space Research Association, Washington DC
Received: 4 August 1992; Accepted: 17 September 1992
Abstract
The Oriented Scintillation Spectrometer Experiment (OSSE) on the Arthur Holly
ComptonGammaRayObservatorysatelliteusesfouractively-shieldedNaI(Tl)-CsI(Na)
phoswich detectors to provide gamma-ray line and continuum detection capability
in the 0:05(cid:0)10 MeV energy range. The instrument includes secondary capabilities
for gamma-ray and neutron detection between 10 and 250 MeV. The detectors have
(cid:14) (cid:14)
3:8 (cid:2)11:4 (FWHM)(cid:12)elds-of-view de(cid:12)ned bytungsten collimators. Each detectorhas
an independent, single-axis orientation system which permits o(cid:11)set pointing from the
spacecraft Z-axis for background measurements and multi-target observations. The
instrument, its calibration and performance are described.
1 INTRODUCTION
The Oriented Scintillation Spectrometer Experiment (OSSE) is one of four experiments
on NASA’s Arthur Holly Compton Gamma Ray Observatory (GRO) satellite. Launched on
1991 April 5, GRO has a complementof instruments which provide coordinated observations
spanning six decades of energy in the gamma-ray range. OSSE has been designed to under-
take comprehensive gamma-ray observations of astrophysical sources in the 0.05 to 10 MeV
energy range. The instrument includes secondary capabilities for gamma-ray and neutron
observations above 10 MeV that will be of particular value for solar (cid:13)are studies.
OSSE follows several previous satellite instruments dedicated to investigations in the
low-energy gamma-ray region, including the gamma-ray spectrometers on the HEAO-1 and
HEAO-3 missions and the gamma-ray spectrometer on the Solar Maximum Mission. These
instruments made pioneering observations which include the detection of di(cid:11)use positron
26 56
annihilation radiation and Al emission from the galactic plane, the detection of Co emis-
sion from SN1987a, the detection of a number of galactic and extragalactic sources up to a
few hundred keV, and the detection of a large number of solar (cid:13)are and cosmic gamma-ray
bursts.
The mission of OSSE is to extend these observations with signi(cid:12)cantly improved sensi-
tivity. OSSE’s observational program addresses a broad range of scienti(cid:12)c objectives in the
following areas:
1. Heavy element nucleosynthesis through observation of radioactivity in supernova rem-
nants,
2. The power source in novae from observations of the associated radioactivity,
3. Binary systems containing neutron stars and black holes,
4. Pulsed and steady-state emissions from pulsars,
5. The di(cid:11)use emissions from the galactic plane and the galactic center regions and pos-
sibly the cosmic di(cid:11)use background,
6. The energy source of active galactic nuclei,
7. The intensity of low-energy cosmic rays and the matter density in the interstellar
medium, and
8. Gamma-ray and neutron emissions from solar (cid:13)ares.
The OSSE instrument was designed to have the capabilities to address the objectives
listed above by adopting a conventional scintillation spectrometer with a rectangular (cid:12)eld-
of-view (FOV), which permits mapping the di(cid:11)use emission along the galactic plane with
modest angular resolution while maintaining the capability to study discrete sources. In
section 2 of this paper the instrument and its capabilities are described. Section 3 presents
the GRO observation constraints and their impact on OSSE observations. In section 4 we
describe the pre-(cid:13)ight OSSE calibration and its results, and the actual on-orbit performance
of OSSE during the (cid:12)rst year of the GRO mission is presented in section 5.
2 INSTRUMENT DESCRIPTION
The OSSE instrument, shown in Figure 1, consists of four identical detector systems
controlled by a central electronics unit. Each of OSSE’s four detectors operates, to a large
degree, as an independent instrument, with separate electronic and pointing systems. The
OSSE detectors are nominally co-aligned with the other pointed GRO instruments, EGRET
and COMPTEL, providing coordinated gamma-ray observations of speci(cid:12)c targets. Each
of the four OSSE detectors can be independently rotated about an axis perpendicular to
the nominal pointing direction of GRO to observe secondary or transient sources such as
the Sun with little impact on spacecraft (S/C) orientation. The characteristics of the OSSE
instrument are summarized in Table 1 and discussed below.
2.1 OSSE Detectors
Figure 2 displays the majorcomponents of oneof thefour OSSE detectors. The primary
detecting element for each is a large area NaI(Tl) scintillation crystal (13-inch diameter by
4-inch thick) shielded in the rear by an optically coupled 3-inch thick CsI(Na) scintillation
crystal in a \phoswich" con(cid:12)guration. Collimation of gamma rays and background reduction
areachievedbytheuseof activeshielding and apassive tungsten collimator. Thephoswich is
enclosed in an annular shield of NaI(Tl) scintillation crystals which provides anticoincidence
for gamma-ray interactions in the phoswich. The annular shield also encloses a tungsten slat
(cid:14) (cid:14)
collimator which de(cid:12)nes the 3:8 (cid:2)11:4 full-width-at-half-maximum (FWHM) gamma-ray
aperture of the phoswich detector. A plastic scintillation detector covers the aperture to
complete a 4(cid:25)-steradian shield for charged particle rejection. The phoswich, annular shield
and associated photomultiplier tubes are completely enclosed in a mu-metal magnetic shield.
2.1.1 Phoswich Detector
The phoswich is viewed from the CsI face by seven 3.5-inch photomultiplier tubes
(PMTs). Inthiscon(cid:12)guration, theCsIportion ofthephoswich actsas anticoincidence shield-
ing for the NaI detector. The detector event processing electronics incorporates pulse-shape
analysis for the discrimination of events occurring in the NaI crystal from those occurring in
the CsI by utilizing the di(cid:11)ering scintillation decay time constants of NaI and CsI.
Table 1: OSSE Characteristics Summary
Detectors
Type: 4 identical NaI-CsI phoswiches,
actively-shielded, passively collimated
2
Aperture Area (total): 2620 cm (geometric)
2
E(cid:11)ective Area (total): 2000 cm at 0.511 MeV (photopeak)
(cid:14) (cid:14)
Field-of-View: 3:8 (cid:2)11:4 FWHM
Energy Resolution: 7.8% at 0.661 MeV
a
3.1% at 6.13 MeV
Time Resolution: 4(cid:0)32 sec in normal mode
0.125 msec in pulsar mode
4 msec in burst mode
Experiment Sensitivities (3(cid:27) for 5(cid:2)105 sec)
0:05(cid:0)10 MeV Line (cid:13)-rays: (cid:24) 2(cid:0)10(cid:2)10(cid:0)5(cid:13) cm(cid:0)2s(cid:0)1
0:05(cid:0)0:5 MeV Continuum: (cid:24) 0:002(cid:2) Crab
Pulsars of known period: (cid:24) 0:003(cid:2) Crab Pulsar
Gamma Ray Bursts: 1(cid:2)10(cid:0)7 erg cm(cid:0)2
Solar Flare Line (cid:13)0s (103 sec (cid:13)are): 1(cid:2)10(cid:0)3 (cid:13) cm(cid:0)2 s(cid:0)1
Solar Flare Neutrons (> 10 MeV): 5(cid:2)10(cid:0)3 n cm(cid:0)2s(cid:0)1
Pointing System
Type: Independent Single Axis
(cid:14)
Range: 192 about the S/C Y-axis
Accuracy: 6 arc minutes
(cid:14)
Speed: 2 /sec (max)
GRO { OSSE Interface
Weight: 1820 kg
Power: 192 watts
Telemetry: 6492 bits/sec
a
Excluding Detector # 4. Detector # 4 has the poorest resolution at all energies, but di(cid:11)ers
signi(cid:12)cantly from the others above (cid:24) 1 MeV.
Energy losses in the phoswich are processed by three separate and independent pulse-
height and pulse-shape analysis systems covering the energy ranges: 0:05(cid:0)1:5 MeV (Low),
1 (cid:0) 10 MeV (Medium), and > 10 MeV (High, nominally 10 (cid:0) 250 MeV). The two lower
ranges are derived from the summed output of the anodes of the seven 3.5-inch PMTs (RCA
83013F). Thehighest rangeis derivedfromthesummedoutput oftheeighth dynodesofthese
PMTs. The pulse-shape discrimination in the highest range is also used to separate neutron
and gamma-ray energy losses in the NaI portion of the phoswich. This discrimination uses
the di(cid:11)ering time characteristics of the secondary particles produced by these interactions
(Share et al. 1978). Gamma-ray event validation in all three ranges includes
1. anticoincidence with energy losses in the NaI annular shield,
2. anticoincidence with aperture charged particle detector, and
3. pulse shape quali(cid:12)cation as energy loss in the phoswich NaI crystal.
The pulse-shape analysis uses rise-time to zero-crossing time measurement on the bipolar-
shaped signals from the PMTs. A time-to-amplitude converter(TAC) measures the shape of
the pulses. Two amplitude discriminators operating on the TAC output de(cid:12)ne a \time win-
dow" which quali(cid:12)es eventsas having shapes consistent with NaI energy losses and validates
them for digitization. Validated events are digitized by 256-channel Wilkinson run-down
analog-to-digital converters. Both pulse height (energy loss) and pulse shape (decay time)
are digitized. The digitized pulse shapes are used for further event quali(cid:12)cation in the form
of energy-dependent pulse-shape discrimination. This energy-dependent discrimination per-
mitstheoptimumrejectionof eventswith partial CsIenergylosses (e.g. Compton scattering)
at each energy. Fully quali(cid:12)ed events are passed on to the spectral accumulation and pulsar
analysis circuits described below. The calibration and on-orbit performance of the pulse
shape discrimination is addressed in section 4 (see Figures 9 and 10).
2.1.2 AGC System and Gain Stability
Optimum spectral resolution in the phoswich detector requires precise gain adjustment
of the seven PMTs viewing the phoswich. This gain balance among the tubes is preservedin
orbit by individual control of each PMT’s high voltage. The high voltage control is provided
by an automatic gain control (AGC) system utilizing a light emitting diode (LED) which
is optically coupled to the CsI part of the phoswich. The signal from each of the PMTs
coincident with the LED (cid:13)ashes is compared with a reference voltage. Di(cid:11)erences between
the reference and the observed PMT signal are used to adjust the high voltage applied to
that PMT. The amplitude of the LED (cid:13)ash is stabilized in a similar way using a PIN diode
which is optically coupled to the LED assembly. This secondary control provides stable LED
output for the PMT gain measurements. The PIN diode system can also be used to change
the amplitude of the LED (cid:13)ash, which in turn changes the gains of the seven phoswich
PMTs while maintaining the gain balance for the detector. Gain increases by up to a factor
of four are possible. The detectors have occasionally been operated in a (cid:2)2 gain mode when
improved spectral performance at the lowest energies is desired (see section 4).
Thespeci(cid:12)ctaskoftheAGCsystemistopreservetherelativegain ofthesevenphoswich
PMTs with respect to temperature and magnetic (cid:12)eld variations. Its ability to preserve the
absolute gain of the phoswich is limited by the temperature dependence of the light output
from the scintillation crystals and the temperature dependence of the spectral response of
(cid:14)
the PMTs. An absolute gain stability of 0.2% per C has been demonstrated in calibrations
(section 4). Due to the very small anticipated detector temperature variation on orbital
(cid:14)
timescales (< 0:1 C), the absolute gain of the OSSE phoswich is expected to be stable to
within 0.1% on timescales of several hours (see section 5).
Since o(cid:11)set pointing is used to measure the OSSE background, the varying magnetic
(cid:12)eld orientations for the target and background pointings could produce di(cid:11)ering gains for
the measurements. The OSSE design has addressed this magnetic (cid:12)eld sensitivity of the gain
of the phoswich PMTs by incorporating, in addition to the AGC system, a double co-netic
shield design. Each PMT is individually housed in a mu-metal shield; a primary continuous
mu-metalshieldcompletelyenclosesthephoswich, annular shieldand theirassociated PMTs.
This shielding achieves a reduction of the external magnetic (cid:12)eld by a factor of 50 or greater
at the PMTs.
The absolute gain of the phoswich is monitored and reported using an internal radioac-
60 60
tive Co source. A Co-doped plastic scintillation detector and associated 0.75-inch PMT
60
(RCA C83012E) is positioned within the tungsten collimator with the Co source nearest
the phoswich. This con(cid:12)guration provides gamma-ray calibration lines at 1.17 and 1.33
(cid:0)
MeV,and a sum peak at 2.5 MeV.By using the coincident (cid:12) energy loss in the plastic scin-
tillator as a calibration event tag, high signal-to-noise calibration spectra are obtained with
relatively weak ((cid:24) 2 nanocurie) radioactive sources (see Figure 19). These 60Co calibration
spectra are routinely collected in a separate data bu(cid:11)er and transmitted in the OSSE data.
2.1.3 Anticoincidence Elements
The NaI(Tl) annular shield enclosing the phoswich is 3.35 inches thick and 13.3 inches
long. It is divided into four optically-isolated quadrants or segments, each viewed by three2-
inch PMTs (RCA S83019F). Energy losses above 100 keVare detectedand used for rejection
of coincident energy losses in the phoswich. Three shield anticoincidence discrimination
signals are used in the event processing:
Low: A 100 keV threshold (programmable from 30 - 470 keV) is used as the low level dis-
criminator for the phoswich low and medium energy ranges,
Med: A1.2 MeVthreshold is used as the low leveldiscriminator for thephoswich high energy
range, and
High: An 8 MeV threshold is used as the upper level discriminator for all phoswich ranges
and triggers a longer rejection pulse for possible cosmic ray overloads.
Annular shield energy losses in the 30 keV to 8 MeV range are pulse-height analyzed as part
of thediagnostic calibration analysis systemdiscussed below. Thegood spectral resolution of
the shield segments (9(cid:0)10% at 0.66 MeV, see section 4) permits the use of shield spectra in
the study of solar (cid:13)ares and gamma-ray bursts. The low level discriminator event rates from
the shields are processed by the OSSE central electronics for the detection of gamma-ray
bursts (see below).
Theannular shield also encloses thetungsten slat collimator which de(cid:12)nesthe detector’s
(cid:14) (cid:14)
rectangular (3:8 (cid:2)11:4 FWHM) (cid:12)eld-of-view. Each collimator was machined from 13-inch
00 00
diameter by 7.25-inch thick sintered ingot of tungsten alloy to form 0:54 (cid:2)1:58 aperture
00 (cid:14) 00 (cid:14)
cells. The tungsten slats are 0:075 thick in the 3:8 direction and 0.113 thick in the 11:4
direction. As discussed in section 4, the high stopping power of tungsten provides passive
collimation which opens only slightly at higher energies.
The aperture of each detector is covered by a charged particle detector (CPD) to com-
plete the 4(cid:25)-steradian charged particle rejection for the phoswich. The CPD consists of a
20-inch square by 0.25-inch thick plastic scintillator (Bicron BC408) viewed by four 2-inch
PMTs (RCA S83019E). Energy losses above a nominal threshold trigger rejection of coinci-
dent energy losses in the phoswich. The threshold is controlled by command but is typically
set at 40% of the energy loss of minimum ionizing protons to provide better than 99.9%
particle detection e(cid:14)ciency.
2.1.4 Detector System Deadtime
The detector is inhibited from accepting new phoswich events during the processing
of previously validated events and during time periods coincident with energy losses in the
anticoincidence elements above their veto thresholds. Livetime-corrected rates in the OSSE
detectors require measurement of this deadtime. Each of the three phoswich energy ranges
has its own deadtime or inhibit signal which is formed from the logical "or" of the phoswich
event processing signals and the veto signals from the anticoincidence elements. The con-
(cid:12)guration of anticoincidence elements contributing to each deadtime signal and their veto
duration can be changed by command. Table 2 summarizes the deadtime elements and their
Table 2: Event Processing Deadtimes
Detector Deadtime per event
Element in element ((cid:22)sec)
Phoswich
Phoswich LLD 3.7
Phoswich ULD 30
Linear Gate 3.5
Conversion 2:5(cid:0)11
Dig Time Window 12
Store 11
a
Anticoincidence
NaI Shield LLD 2.3 / 5
b
NaI Shield MLD 2.3 / 5
NaI Shield ULD 10 / 30
CPD LLD 2.3 / 5
a
Shield and CPD are command controlled to one of two veto times.
b
Medium Level Discriminator, (cid:24) 1:2 MeV for phoswich high range
duration. Thedeadtimemeasurementconsists ofsampling theinhibit signal atafrequencyof
128 kHz; a deadtime counter is incremented for each sample which indicates that the system
is inhibited. The deadtime counter is transmitted in the detector housekeeping information
with a time resolution of 250 (cid:22)sec which provides typical accuracy of (cid:1)T/T (cid:24) 2(cid:2)10(cid:0)5.
The phoswich event processing consists of 1) amplitude quali(cid:12)cation, 2) shape quali-
(cid:12)cation, 3) conversion, 4) digital time window quali(cid:12)cation, and 5) memory storage. The
conversion is performed bya Wilkinson run-down system which has an amplitude-dependent
duration. The worst case phoswich event processing time is (cid:24) 42 (cid:22)sec, but the digital time
window and storage processing run independently of the quali(cid:12)cation process. Thus the
phoswich deadtime per quali(cid:12)ed event is in the range (cid:24) 10 (cid:0) 18 (cid:22)sec. Phoswich events
which exceed the energy range upper level discriminator (ULD) are rejected and have a
(cid:12)xed deadtime of 30 (cid:22)sec. The anticoincidence elements contribute a (cid:12)xed deadtime per
event above threshold. The time interval is either 2.3 or 5 (cid:22)sec as selected by command.
The shield upper level discriminator veto durations are 10 or 30 (cid:22)sec.
